The objective of this research is to determine the functional organization of normal and pathologic motor pattern generation in humans in vivo with positron emission tomography imaging. Contrasting paradigms requiring locomotion, visual pursuit, and prehension will be studied with PET images of blood flow and glucose metabolism. We predict that a given type of patterned motor behavior will generate a specific distribution of active cerebral areas that can be used to define functional connectivity. We hypothesize that motor control can be effectively modelled with PET by characterizing the relative hierarchy of command centers and the degree to which they are distributed as a network of parallel cortical and subcortical processors. Cerebral plasticity associated with motor command centers. Measurable changes would include relocation of sites of cerebral response, alterations in the magnitude of each response, or redistribution of the functional connectivity between sites. Findings in normal subjects will form a basis for the study of CNS pathology. Locomotion, motor learning, and prehension will be examined in Parkinson's disease patients to delineate the metabolic anatomy of adaptation in response to a neurodegenerative process. Longitudinal studies of motor control in patients recovering from stroke will also be examined to characterize subacute functional reorganization of the motor system. By characterizing the potential configurations of connected and distributed networks of local neural responses that are active during motor performance, we will provide a broad view of motor organization. This view will have a significant impact on our understanding of the functional, as opposed to anatomic, defects that underlie nervous system injury and degeneration. Knowledge of the dynamic changes of the motor system during functional reorganization may ultimately guide therapy and predict outcome.